Examining the Levallois Reduction Strategy From a Design Theory Point of ViewDennis M. SandgatheOxford: John and Erica Hedges Ltd., BAR International Series 1417, 2005, 176 pp. (paperback), 41.00.ISBN 1-84171-717-7

Reviewed by: IAN J. WALLACEInterdepartmental Doctoral Program in Anthropological Sciences, Stony Brook University, Stony Brook, NY 11794-4364, USA

The Levallois reduction strategy has long been consid-ered a hallmark of the Middle Paleolithic of Western Eurasia (Bordes 1953) and the Middle Stone Age of Africa (Goodwin 1929). However, both the definition of the tech-nique and its possible significance remain controversial pa-leoanthropological issues (see papers in Dibble and Bar-Yo-sef 1995). Sandgathes volume, based on his doctoral thesis, represents a recent contribution to the ongoing Levallois discussion.

This study attempts to elucidate the potential advan-tages of employing the Levallois reduction strategy, which contributed to its widespread geographic distribution and temporal endurance. Constructed around design theory, the study develops a generalized model of Middle Paleo-lithic lifeways and attempts to identify those factors that would have most significantly constrained the design of lithic technologies. These constraints take on two formsfunctional constraints on the flaking products (blanks and tools) and those imposed on the reduction sequence. Five morphological aspects of blanks are hypothesized to more or less dictate performanceflake size, length of cutting edge, angle of cutting edge, ventral curvature, and over-all robusticity. Raw material economy and technological knowledge are proposed as the primary constraints influ-encing the choice of reduction strategy. Based on these con-straints, several hypotheses are developed that are argued to potentially explain the advantages of employing a Leval-lois reduction strategy and the circumstances under which these operate.

These hypotheses are tested through the analysis of blank selection patterns for tool-making at three French Middle Paleolithic sites in the Southwest (Pech de LAz IV, le Moustier, and Combe Capelle Bas) and one in the Southeast (Jiboui). The analysis begins by comparing the attributes thought to control performance in the morpholo-gies of used vs. unused flakes. Sandgathe concludes that selection, in fact, does appear to have been based on all of the hypothesized attributes except ventral curvature. In ad-dition, the selection of specific attributes varied depending on the conditions of the site.

The analysis then considers how blanks produced by various reduction strategies from each of the assemblag-es might satisfy the selection criteria that appear to have been in place. The four classes of blanks considered include central Levallois flakes, peripheral Levallois flakes, clats dbordants, and amorphous flakes. By comparing the char-acteristics of the technological products to the characteris-tics of those blanks chosen for use, Sandgathe concludes that Levallois products, as well as the products of similar

reduction approaches, possess functionally advantageous morphologies. The most significant attributes appear to be greater average size, a high number of usable edges per flake, and decreased robusticity. A more general conclusion of the study is that Levallois technologies would offer im-portant advantages under conditions of restricted access to raw material.

The basic questions explored in this study are rooted in evolutionary ecology in that they seek to investigate the ways in which the adoption of the Levallois technique can be considered an adaptive strategy pursued by Middle Pa-leolithic humans. In this sense, the research employs high-level theory that is closely aligned with an Americanist approach to lithic technology. Likewise, one of the great strengths of the work that Sandgathe presents in this book and elsewhere (Sandgathe 2004) is that it approaches Mid-dle Paleolithic core reduction strategies from the perspec-tive of the overall organization of technology (sensu Nelson 1991). Specifically, this is achieved by expanding the focus of investigation beyond classic Levallois products (sensu Bordes 1961) to include the significance of products such as clats dbordants. In fact, this book may represent the most thorough investigation of the functionality of clats dbor-dants yet published.

Most of the problems from which this study suffers, however, relate to the precision with which types of flaking products (i.e., blanks) are defined and the degree to which they can be assigned accurately to specific reduction strate-giesissues that have been the focus of considerable ar-chaeological debate (e.g., Boda 1994: 262-265; Dibble 1989; Marks and Volkman 1983). Because blanks are the studys primary unit of analysis, these obstacles could present sig-nificant challenges to the conclusions. Two of the four blank types considered by Sandgathe are particularly problem-atical and deserve some discussion. These are clats dbor-dants and peripheral Levallois flakes.

Throughout the book, Sandgathe maintains that clats dbordants are widely recognized as a product of discoidal technologies. However, many key studies of Levallois tech-nology describe clats dbordants as a far more general class of blanks which can be produced by both discoidal and Levallois techniques (Boda 1988, 1993, 1994; Boda et al. 1990; Debnath and Dibble 1994; Meignen 1995). A simple definition of these artifacts could be flakes whose lateral margin removes an edge of the core. Because a techno-logical understanding of this blank type is necessary to fol-low the logic of several of the arguments being presented throughout the book, it would have been better if Sand-gathe had made more explicit his reasoning for associating

PaleoAnthropology 2006: 5153. Copyright 2006 Paleoanthropology Society. All rights reserved.

52 PaleoAnthropology 2006

clat dbordants with a discoidal reduction system. A dia-gram (Figure 2.4) tries to clarify this somewhat, but it is not effective. Illustrations of the artifacts that were classified as clats dbordants following standard conventions (sensu Addington 1986) might have helped clarify this issue, but only undiagnostic photographs are provided (Plates 1-3).

Peripheral Levallois is a second tenuous class of blanks considered in the study. Sandgathe defines these in the fol-lowing way:

If its shape and dorsal scar configuration indicate that it comes from the periphery of the face of the core and it includes a significant portion of the peripheral edge of sub-circular shaped, single-surface core then it is consid-ered a centripetal or peripheral Levallois flake (see Van Peer 1992) or potentially a more generic disc core flake (p. 83).

Here again, as was the case with the clats dbordants class, this type potentially includes both Levallois and dis-coidal products. As well, this class would seem to include several forms of clats dbordants.

These techno-typological issues limit the study in that they prohibit the examination of Levallois core reduction as a discrete phenomenon, which is a stated objective of the study. Instead, what is actually being examined is a greater class of blanks which includes the products of all variants of Levallois and discoidal technologies. While there may be some evidence that these technologies are related (Baum-ler 1988; Lenoir and Turq 1995), many researchers believe that they are technologically divergent enough to warrant individual consideration. This is not only because they are organized differently volumetrically (Boda 1993), but also because they represent different levels of curation and are likely to have different costs and benefits (Brantingham and Kuhn 2001; Dibble 1997). Regardless of what position one takes on the relatedness of Levallois and discoidal tech-nologies, the system employed to type blanks in this study clearly masks potentially significant technological variabil-ity. Moreover, while one of the primary conclusions of the book is that several of the advantageous morphological as-pects of Levallois products are found in similar technolo-gies, this seems to be more of an assumption inherent in the system used to type blanks than it is an actual finding.

Other problems arise from the basic methodology of design theory itself. First, the identification of possible con-straints on lithic technologies, which serve as the founda-tion for the hypotheses tested in the study, is a function of what is known about Middle Paleolithic human adapta-tions. Likewise, Sandgathes predictions about optimal tool design rely almost entirely on taphonomically complicated use-wear data which indicate that stone tools were primar-ily used in the Middle Paleolithic for woodworking and butchery. However, these are the same conclusions reached by use-wear studies of Lower Paleolithic assemblages where Levallois was not common (Shea 2006). Therefore, these use-wear data do not inform on how Levallois products can be considered uniquely designed for these purposes.

Second, even if it were possible to unequivocally as-

sign a flake exhibiting use-wear to a certain technology, this does not mean that they were actually designed for that purpose. This would seem to be supported by the fact that among each of the blank classes analyzed by Sandgathe, the overall frequency of used and/or retouched amorphous blanks was higher than any other blank class (pp.126130). Importantly, selection of a flake for retouch or use can take place long after a flake was produced, to judge from the not infrequent occurrence of retouch scars bearing younger patination than the remainder of the flake. Therefore, while one of the primary goals of the study is to evaluate the adaptive significance of the Levallois reduction strategy, the behavior being examined is simply blank selection pat-terns. To consider this aspect of behavior indicative of the entire flintknapping strategy is to reify the existence of a desired end-product, which may or may not be an emic goal regardless of any etic validity (i.e., the Ford (1952) vs. Spaulding (1953) debate).

Third, in assessing the role of sociality in Levallois technological knowledge, Sandgathe seems to discount the role of social learning. This conclusion runs counter to an enormous body of ethnographic data on how individuals acquire technical skills (e.g., Wiessner 1983, 1984; see also references in Shennan and Steele 1999) and even studies of modern flintknappers (Whittaker 2004) suggesting that complex tasks (such as the production of Levallois flakes) are most effectively transmitted in social contexts.

There are also issues with the overall production quality of the book. The bibliography could have been more thor-ough, particularly for work done by French technologists. Specifically, strategic placement of such citations may have helped clarify the source of Sandgathes understanding of clats dbordants. As well, while the book contains numer-ous graphs and tables which are all very clear and rich with data, schematic representations of reduction sequences and the photographs of the artifacts are unclear. And as previ-ously mentioned, the book also might have benefited from the inclusion of artifact illustrations. Typographical errors exist, but none compromise the readability.

Despite these criticisms, readers interested in Levallois technology are likely to find many of the ideas presented by Sandgathe to be insightful and thought-provoking. Because none of the assemblages considered in the study have any significant Levallois point component, it is uncertain if the findings would be germane to work being done in regions such as the Levant where there are higher frequencies of such artifacts during the Middle Paleolithic than there are in Western Europe. The book additionally will be of inter-est to those researchers more generally studying Middle Paleolithic stone tool technology.

REFERENCESAddington, L.R. 1986. Lithic Iillustration: Drawing Flaked

Stone Artifacts for Publication. Chicago: University of Chicago Press.

Baumler, M. 1988. Core Reduction, Flake production, and the Middle Paleolithic Iindustry of Zobite (Yugosla-via). In Upper Pleistocene of Western Eurasia, H. Dibble

BOOK REVIEW 53

and A. Montet-White (eds.), pp. 255274. Philadelphia: University of Pennsylvania Press.

Boda, . 1988. Le concept Levallois et valuation de son champ dapplication. In LHomme de Nandertal, 4: La technique, tudes et recherches archologiques de luniversit de Lige 31, M. Otte (ed.), pp. 1326. Lige.

Boda, . 1993. Le dbitage discode et le dbitage rcur-rent centripte. Bulletin de la Socit Prhistorique Fran-aise 90: 392404.

Boda, . 1994. Le concept Levallois: Variabilit des methods. Monographie du CRA 9. Paris: CNRS ditions.

Boda, ., Geneste, J.-M., and L. Meignen. 1990. Identifica-tion de chanes opratoires lithiques du Palolithique ancien et moyen. Palo 2: 4380.

Bordes, F. 1953. Levalloisien et Moustrien. Bulletin de la So-cit Prhistorique Franaise 50: 226235.

Bordes, F. 1961. Typologie du Palolithique ancien et moyen. Bordeaux: Demlas.

Brantingham, P.J., and S.L. Kuhn. 2001. Constraints on Lev-allois Core Technology: A Mathematical Model. Journal of Archaeological Science 28: 747761.

Debnath, A., and H.L. Dibble. 1994. Handbook of Paleolithic typology, Vol. 1: Lower and Middle Paleolithic of Europe. Philadelphia: University of Pennsylvania Press.

Dibble, H.L. 1989. The Implications of Stone Tool Types for the Presence of Language During the Lower and Mid-dle Paleolithic. In The Human Rrevolution: Behavioral and Biological Perspectives on the Origins of Modern Humans, P. Mellars and C. Stringer (eds.), pp. 415432. Prince-ton: Princeton University Press.

Dibble, H.L. 1997. Platform Variability and Flake Morphol-ogy: A Comparison of Experimental and Archaeologi-cal Data and Implications for Interpreting Prehistoric Lithic Technological Strategies. Lithic Technology 22(2): 150170.

Dibble, H.L., and O. Bar-Yosef (eds.). 1995. The Definition and Interpretation of Levallois Technology. Madison: Pre-history Press.

Ford, J.A. 1952. Measurements of Some Prehistoric Design De-velopments in the Southeastern States. American Museum of National History, Anthropological Papers 44 (3): 313384.

Goodwin, A.J.H. 1929. The Middle Stone Age. Annals of the South African Museum 27: 95145.

Lenoir, M., and A. Turq. 1995. Recurrent Centripetal Deb-itage (Levallois and Discoidal): Continuity or Discon-tinuity? In The Definition and Interpretation of Levallois Technology, H.L. Dibble and O. Bar-Yosef (eds.), pp. 249256. Madison: Prehistory Press.

Marks, A. E., and P. Volkman. 1983. Changing Core Reduc-tion Strategies: A Technological Shift from the Middle to Upper Paleolithic in the Southern Levant. In The Mousterian Legacy: Human Biocultural Change in the Up-per Pleistocene, E. Trinkaus (ed.), pp. 1334. Oxford: BAR International Series.

Meignen, L. 1995. Levallois Lithic Production Systems in the Middle Paleolithic of the Near East: The Case of the Unidirectional Method. In The Definition and Interpreta-tion of Levallois Technology, H.L. Dibble and O. Bar-Yosef (eds.), pp. 361379. Madison: Prehistory Press.

Nelson, M. 1991. The Study of Technological Organization. In Archaeological Method and Ttheory, Vol. 3, M. Schiffer (ed.), pp. 57100. Tucson: University of Arizona Press.

Sandgathe, D.M. 2004. An Alternative Interpretation of the Levallois Technique. Lithic Technology 29(2):148159.

Shea, J.J. 2006. What Stone Tools Can (and Cant) Tell Us About Early Hominin Diets. In Early Hominin Diets: The Known, the Unknown and the Unknowable, P. Ungar (ed.), pp. 321351. Oxford: Oxford University Press.

Shennan, S., and J. Steele. 1999. Cultural Learning in Homi-nids: A Behavioural Ecological Approach. In Mamma-lian Social Learning: Comparative and Ecological Perspec-tives, H.O. Box and K.R. Gibson (eds.), pp. 367388. Cambridge: Cambridge University Press.

Spaulding, A.C. 1953. Statistical Techniques for the Discov-ery of Artifact Types. American Antiquity 18: 305313.

Van Peer, P. 1992. The Levallois Reduction Strategy. Madison: Prehistory Press.

Whittaker, J.C. 2004. American Flintknappers: Stone Age Art in the Age of Computers. Austin: University of Texas Press.

Wiessner, P. 1983. Style and Social Information in Kalahari San Projectile Points. American Antiquity 48: 253276.

Wiessner, P. 1984. Reconsidering the Behavioral Basis for Style: A Case Study Among the Kalahari San. Journal of Anthropological Archaeology 3: 190234.